Skip to content
Back to search

agent-toolkit

MarketplaceWorks with Claude CodeGitHub

A collection of project-agnostic skills for common engineering tasks such as ticket refinement, planning, code reviews, agent memory hygiene, skill authoring, and more.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today

Check it deeper on your machine

It has changed since its scan on 11 Nov 2026.

As a Claude Code plugin, install it and then ask Claude Code:

/plugin marketplace add keithmackay/mackayi/plugin install scanskillsafety@mackayiDo a deep scan of https://github.com/eai-org/agent-toolkit with scanskillsafety

Or from a terminal, clone it and run it:

git clone https://github.com/keithmackay/scanskillsafety-skill ~/.claude/skills/scanskillsafetynode ~/.claude/skills/scanskillsafety/cli.cjs --deep 'https://github.com/eai-org/agent-toolkit'

It checks the exact commit: every file, its packages against OSV (api.osv.dev), how it compares with this page's scan (findsafeskills.com receives the repository's address), and, if you agree, Claude reading every file with your own Claude subscription. It never runs the code.

Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/eai-org--agent-toolkit/badge.svg)](/listings/eai-org--agent-toolkit)

HTML

<a href="/listings/eai-org--agent-toolkit"><img src="/api/listings/eai-org--agent-toolkit/badge.svg" alt="findsafeskills safety rating" /></a>

Install

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

Install all skills in one command:

git clone https://github.com/eai-org/agent-toolkit.git && cd agent-toolkit && ./install.sh

Update in one command:

cd agent-toolkit && git pull && ./install.sh

On Claude Code the install also registers a hook that runs that update for you once a day; an install made before it needs the command above once to pick it up ([docs/auto-update.md](./docs/auto-update.md)).

Every skill is installed by default. To leave one out, name it once and the choice sticks, daily update included:

./install.sh --exclude use-conversational-language

`--include <name>` puts it back. Both are repeatable.

On Windows the skills link through junctions, so the update command above works as-is, though an install predating them needs one `./install.sh --force` to convert. The rules are copied instead and need `--force` every time (see [Windows](./docs/install-skills.md#windows)).

How the symlink install works and the other install methods — hand-picking skills, other agents, [skills.sh](https://skills.sh/), the Claude Code plugin marketplace — are covered in [docs/install-skills.md](./docs/install-skills.md).

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

.claude-plugin/marketplace.json · Manifest · 11 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

This file has changed since it was scanned, so this is not exactly the text that was scanned, and no line is marked. The findings are listed with the safety scan above.

  1. {
  2. "name": "agent-toolkit",
  3. "owner": {
  4. "name": "EAI",
  5. "email": "engineering.age.ai@gmail.com"
  6. },
  7. "description": "A collection of project-agnostic skills for common engineering tasks such as ticket refinement, planning, code reviews, agent memory hygiene, skill authoring, and more.",
  8. "plugins": [
  9. { "name": "agent-toolkit", "source": "./" }
  10. ]
  11. }
README.md · README · 360 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

This file has changed since it was scanned, so this is not exactly the text that was scanned, and no line is marked. The findings are listed with the safety scan above.

  1. # agent-toolkit
  2. A collection of generic agentic tools for common engineering tasks, designed to work with any AI
  3. agent on any kind of software project. The philosophy behind the toolkit — five pillars and a
  4. litmus test for every new artifact — is captured in
  5. [docs/core-philosophy.md](./docs/core-philosophy.md).
  6. ## Skills
  7. Agentic skills for everyday software engineering — solo or in a team, not bound to any specific
  8. language or framework.
  9. ### Skill & doc authoring
  10. Tools to create and continuously improve the skills and docs your agents rely on, following
  11. [this approach to agentic skills](https://medium.com/engineering-in-the-age-of-ai/my-approach-to-agentic-skills-e08dc6c0d1cd).
  12. - **[compact-docs-writer](./skills/compact-docs-writer/SKILL.md)** — write docs with maximum token
  13. economy.
  14. - **[compact-skill-creator](./skills/compact-skill-creator/SKILL.md)** — create or edit skills,
  15. keeping them lean and efficient.
  16. - **[self-improve](./skills/self-improve/SKILL.md)** — capture a lesson into the skill, doc or check
  17. that governs it, so mistakes aren't repeated and agents keep getting better at the project.
  18. ### Context & memory hygiene
  19. Maintenance to run from time to time, keeping your setup tidy and your context sharp.
  20. - **[agentify-project](./skills/agentify-project/SKILL.md)** — make a project agent-ready by
  21. converging it toward a canonical agent-neutral structure (a lean AGENTS.md index with
  22. progressive disclosure, shared skills, gitignore hygiene). Re-runnable, and doubles as an audit.
  23. - **[context-checkup](./skills/context-checkup/SKILL.md)** — audit what auto-loads into a
  24. session's context and spot what can be trimmed to reduce startup tokens.
  25. [Why this is important](https://medium.com/engineering-in-the-age-of-ai/keep-your-ai-agents-context-window-sharp-7255d83a8949).
  26. - **[memory-doctor](./skills/memory-doctor/SKILL.md)** — drain the memory your agents keep
  27. auto-accumulating, block by block, moving each entry to the right place or archiving it.
  28. [More about memory-doctor](https://medium.com/engineering-in-the-age-of-ai/keep-your-ai-agents-memory-clean-and-organized-with-memory-doctor-a79f7174f257).
  29. ### Task workflow
  30. A daily routine for any programming task, following the
  31. [RPAC workflow](https://medium.com/engineering-in-the-age-of-ai/the-refine-plan-act-pattern-for-agentic-ai-coding-59ee013e4427):
  32. fetch a ticket, refine it, harden it, plan it, harden it again, split the plan into small tasks,
  33. then let a fresh session execute them one at a time and consolidate the changes.
  34. - **[fetch-ticket](./skills/fetch-ticket/SKILL.md)** — download a ticket from any tracker
  35. (e.g. GitHub, Jira, Azure DevOps) and save it as a self-contained markdown file.
  36. - **[attach-to-ticket](./skills/attach-to-ticket/SKILL.md)** — save a pasted screenshot, a file
  37. or a URL beside a fetched ticket and reference it from the ticket file, e.g. to fill an
  38. attachment the fetch could not download.
  39. - **[grill-me](./skills/grill-me/SKILL.md)** — get interviewed relentlessly about a plan, design
  40. or idea, one question at a time, each with a recommended answer, until every branch of the
  41. decision tree is resolved; ends with a paste-ready list of the settled decisions. Adapted from
  42. [Matt Pocock's grill-me](https://github.com/mattpocock/skills) (MIT, copyright 2026 Matt Pocock).
  43. - **[refine-ticket](./skills/refine-ticket/SKILL.md)** — define the "what" of a task: validate the
  44. ticket — or a raw idea you want to brainstorm — against the codebase, settle open decisions
  45. together, and save a self-contained requirements doc a fresh session can pick up. Takes a
  46. fetched ticket file, or a ticket URL a subagent fetches first.
  47. - **[create-implementation-plan](./skills/create-implementation-plan/SKILL.md)** — define the
  48. "how" of a task: turn the requirements into an implementation plan, settling the technical
  49. decisions together, then save it for a fresh session to execute.
  50. - **[harden-artifact](./skills/harden-artifact/SKILL.md)** — harden a requirements doc or a plan
  51. before the next phase picks it up: a fresh-context reviewer tests every deviation from the
  52. ticket and every claim against the code and related tickets, the session challenges each
  53. finding, and proven mistakes get fixed in the document (say `autoaccept` to skip the asks for
  54. those). Findings that reveal a durable lesson end in an offer to run self-improve, when
  55. installed.
  56. - **[split-plan-tasks](./skills/split-plan-tasks/SKILL.md)** — break a plan into individually
  57. reviewable tasks grouped into PR-sized batches, appended to the plan file for a fresh session to
  58. execute one at a time.
  59. [Why not execute the full plan at once](https://medium.com/engineering-in-the-age-of-ai/do-not-execute-the-full-plan-break-it-down-step-by-step-2e0ba30ddf73).
  60. - **[execute-plan-tasks](./skills/execute-plan-tasks/SKILL.md)** — execute one task of that
  61. breakdown at a time: implement it, verify it, tick it off in the plan, and hand it back for
  62. review before the next; when a group is done, hand it to self-review (if installed) before the
  63. next group starts. A task whose approach fails twice, or needs reshaping, stops unticked with a
  64. hand-off to re-plan or harden the plan (if installed).
  65. - **[create-manual-test-instructions](./skills/create-manual-test-instructions/SKILL.md)** —
  66. derive manual test steps from a ticket or requirements file, or from a code change (branch,
  67. PR, commit range, diff), useful for the developer or QA.
  68. - **[handover](./skills/handover/SKILL.md)** — package a finished change for its reviewers:
  69. what it does and why, the decisions and plan deviations worth knowing, a review guide, and a
  70. diagram or tree wherever the diff alone hides the change's shape — paste-ready as the PR
  71. description. [More about handover](https://medium.com/engineering-in-the-age-of-ai/the-missing-step-in-agentic-coding-the-handover-d1963c3d2c1d).
  72. - **[prepare-prompt](./skills/prepare-prompt/SKILL.md)** — write the prompt that hands this
  73. session's work to a fresh session — what exists, what is open, what to do and in what form, the
  74. facts it cannot derive — as short as the next session allows.
  75. ### Review assistants
  76. Review helpers that check the codebase while assisting with code or ticket reviews.
  77. - **[fetch-pr-review](./skills/fetch-pr-review/SKILL.md)** — collect the comments left by other
  78. reviewers on your PR and save them into a markdown doc, pinned to the PR head, ready to address
  79. (or push back on), for example via refine-pr-review.
  80. - **[refine-pr-review](./skills/refine-pr-review/SKILL.md)** — go through a fetched PR review
  81. together, comment by comment — address, partial, or push back — drafting the replies, turning
  82. the accepted changes into a requirements doc, and recording the durable lessons, which
  83. self-improve persists when installed; a small change set and the lessons can be handed to
  84. subagents from the same session. Takes the fetched file, or a PR URL a subagent fetches first.
  85. - **[extract-pr-review-rules](./skills/extract-pr-review-rules/SKILL.md)** — mine the conventions
  86. your reviewers actually enforce out of every PR review comment in a repo, triage them with you in
  87. one batch, and persist them through self-improve into the project's topic docs, so each rule
  88. loads only with its subject; later runs resume from a checkpoint and read only the new comments.
  89. - **[review-code-assistant](./skills/review-code-assistant/SKILL.md)** — prepare a review of a PR
  90. or branch with you: the diff the platform shows, the project's conventions run as a checklist,
  91. and candidate comments in diff order, read-only and never posted.
  92. - **[maintainer-review](./skills/maintainer-review/SKILL.md)** — review someone else's PR as the
  93. maintainer deciding whether it merges: every prior comment walked, every claim verified, then
  94. comment, approve or fix the contributor's branch on your go-ahead.
  95. [How open source projects can cope with AI slop](https://medium.com/engineering-in-the-age-of-ai/how-open-source-projects-can-cope-with-ai-slop-9ca7b015602e).
  96. - **[review-ticket](./skills/review-ticket/SKILL.md)** — triage a ticket or ticket set before
  97. anyone picks it up, saving a review with a feature walkthrough and the questions worth raising
  98. with the team.
  99. - **[verify-understanding](./skills/verify-understanding/SKILL.md)** — explain the feature back
  100. in your own words before building it: a teach-back conversation over its ticket review that
  101. probes and corrects until you are ready to implement.
  102. - **[check-ticket-implementation](./skills/check-ticket-implementation/SKILL.md)** — check how
  103. much of a ticket is already implemented in the code, saving a per-requirement status report
  104. (done, partial, not done, or not verifiable).
  105. - **[fresh-eyes-review](./skills/fresh-eyes-review/SKILL.md)** — let an agent with a fresh
  106. perspective review a changeset and report its findings back to the main session. [Read more about fresh-eyes-review](https://medium.com/engineering-in-the-age-of-ai/more-powerful-ai-reviews-with-fresh-eyes-bfad221748c0).
  107. - **[self-review](./skills/self-review/SKILL.md)** — get your changeset merge-ready before
  108. requesting review: a fresh-context reviewer checks it as a maintainer would, you answer every
  109. finding (or say `autoaccept` to let the agent decide), and a compact report for the PR proves
  110. the review happened. Findings that reveal a durable lesson end in an offer to run self-improve,
  111. when installed.
  112. Projects can tune the review via an optional `.agents/docs/self-review-rules.md`.
  113. ### Talking to humans
  114. Wording for what the agent writes to people, whether the user in the session or other humans.
  115. - **[explain-in-simple-language](./skills/explain-in-simple-language/SKILL.md)** — word
  116. explanations, recaps, questions to the user, and documents humans read so they are understood
  117. on the first read, without dumbing them down; voice stays with the caller.
  118. - **[use-conversational-language](./skills/use-conversational-language/SKILL.md)** — the voice for
  119. text that should read as if a person typed it, layered on explain-in-simple-language; used by
  120. the review skills for comments and replies and by rules for texts other people read and code
  121. comments.
  122. ### Code checks
  123. - **[run-nx-checks](./skills/run-nx-checks/SKILL.md)** — run format, lint, typecheck, test, and
  124. build on the affected projects of an Nx workspace and fix unambiguous failures, reporting every
  125. edit it made.
  126. ### How to install the skills
  127. Install all skills in one command:
  128. ```sh
  129. git clone https://github.com/eai-org/agent-toolkit.git && cd agent-toolkit && ./install.sh
  130. ```
  131. Update in one command:
  132. ```sh
  133. cd agent-toolkit && git pull && ./install.sh
  134. ```
  135. On Claude Code the install also registers a hook that runs that update for you once a day; an
  136. install made before it needs the command above once to pick it up
  137. ([docs/auto-update.md](./docs/auto-update.md)).
  138. Every skill is installed by default. To leave one out, name it once and the choice sticks, daily
  139. update included:
  140. ```sh
  141. ./install.sh --exclude use-conversational-language
  142. ```
  143. `--include <name>` puts it back. Both are repeatable.
  144. On Windows the skills link through junctions, so the update command above works as-is, though an
  145. install predating them needs one `./install.sh --force` to convert. The rules are copied instead
  146. and need `--force` every time (see [Windows](./docs/install-skills.md#windows)).
  147. How the symlink install works and the other install methods — hand-picking skills, other agents,
  148. [skills.sh](https://skills.sh/), the Claude Code plugin marketplace — are covered in
  149. [docs/install-skills.md](./docs/install-skills.md).
  150. ## Rules
  151. A set of generic, project-agnostic, opinionated rules that apply to any codebase. They are opt-in,
  152. installed separately from the skills.
  153. - **[compact-governing-docs](./rules/compact-governing-docs.md)** — run the matching compaction
  154. skill before writing or editing a governing doc, so it stays compact.
  155. - **[git-read-only-by-default](./rules/git-read-only-by-default.md)** — never commit, push, merge,
  156. or otherwise write to git without an explicit instruction.
  157. - **[no-ai-attribution](./rules/no-ai-attribution.md)** — no AI co-author trailers, "Generated with"
  158. footers, session links or trailers, or other AI-attribution lines on commits and PRs.
  159. - **[no-nonsense-comments](./rules/no-nonsense-comments.md)** — write only code comments that
  160. still make sense to a future reader with zero context, prefer no comment over a low-value one,
  161. and voice them via [use-conversational-language](./skills/use-conversational-language/SKILL.md).
  162. - **[plans-directory](./rules/plans-directory.md)** — save plans and similar documents under the
  163. project's planning directory, following a certain structure; one per repository, resolved to the
  164. main checkout from a linked worktree.
  165. - **[read-other-repos-governing-docs](./rules/read-other-repos-governing-docs.md)** — before
  166. editing another repo's files, read and follow that repo's governing docs — they don't auto-load.
  167. - **[self-contained-docs](./rules/self-contained-docs.md)** — keep planning and design docs, and
  168. prompts for another session, concise and executable by a fresh session with no prior context,
  169. writing the prompts via [prepare-prompt](./skills/prepare-prompt/SKILL.md).
  170. - **[self-improve-on-correction](./rules/self-improve-on-correction.md)** — when the user corrects
  171. something a skill or doc governs, offer to persist the lesson via
  172. [self-improve](./skills/self-improve/SKILL.md).
  173. - **[write-realistic-texts](./rules/write-realistic-texts.md)** — make text other people will
  174. read sound natural, and get the user's go-ahead on the wording before publishing the texts it
  175. names.
  176. - **[write-simple-explanations](./rules/write-simple-explanations.md)** — word questions,
  177. explanations, recaps to the user, and documents humans read via
  178. [explain-in-simple-language](./skills/explain-in-simple-language/SKILL.md).
  179. One more rule for non-English speakers ships as a copy-paste snippet rather than a file, since it
  180. needs the language filled in — [use-my-mothertongue](./docs/use-my-mothertongue-rule.md) makes
  181. your agent talk to you in your language while everything it writes into the project stays English.
  182. ### How to install the rules
  183. Because rules are always-on, they are never installed implicitly with the skills. To opt in:
  184. ```sh
  185. ./install-opinionated-rules.sh
  186. ```
  187. Auto-loaded rules are mostly a Claude Code feature — how they work, which agents support them, the
  188. script's options, and linking individual rules by hand are covered in
  189. [docs/install-rules.md](./docs/install-rules.md).
  190. ## Install with agentwheel
  191. [agentwheel](https://github.com/NestDevLab/agentwheel) installs this repo's rules **and** skills
  192. into your agent and keeps them in sync across Claude, Codex, Copilot, and other runtimes, from
  193. one source. Run it from where you want it installed (`~` for user level, or a project root):
  194. ```sh
  195. npx agentwheel install github:eai-org/agent-toolkit --adapter claude
  196. ```
  197. Other adapters, selecting individual pieces, and the OpenPack manifest are covered in
  198. [docs/install-with-agentwheel.md](./docs/install-with-agentwheel.md).
  199. ## Artifact relationships
  200. Solid arrows are the `requires` pairs declared in [`openpack.json`](openpack.json); dotted "when
  201. available" arrows are optional siblings a skill uses when present, as a next-phase hand-off or an
  202. in-run call; dotted "informs" arrows are rules shaping a skill's output.
  203. ```mermaid
  204. flowchart TD
  205. fetch_ticket["fetch-ticket"] -. when available .-> refine["refine-ticket"]
  206. fetch_ticket -. when available .-> attach["attach-to-ticket"]
  207. attach -. when available .-> refine
  208. fetch_pr["fetch-pr-review"] -. when available .-> refine_pr["refine-pr-review"]
  209. refine_pr --> refine
  210. refine_pr -. when available .-> plan
  211. refine_pr --> express["use-conversational-language"]
  212. review_code["review-code-assistant"] --> express
  213. review_code --> simple["explain-in-simple-language"]
  214. self_review["self-review"] --> fresh_eyes["fresh-eyes-review"]
  215. self_review -. when available .-> self
  216. simple -. when available .-> self
  217. harden["harden-artifact"] --> fresh_eyes
  218. harden -. when available .-> plan
  219. harden -. when available .-> manual
  220. harden -. when available .-> split
  221. harden -. when available .-> execute
  222. refine -. when available .-> harden
  223. refine -. when available .-> fetch_ticket
  224. refine_pr -. when available .-> fetch_pr
  225. refine_pr -. when available .-> harden
  226. plan -. when available .-> harden
  227. harden -. when available .-> simple
  228. harden -. when available .-> self
  229. maintainer_review["maintainer-review"] --> review_code
  230. maintainer_review --> fresh_eyes
  231. maintainer_review --> express
  232. realistic_rule["write-realistic-texts rule"] --> express
  233. nonsense_rule["no-nonsense-comments rule"] --> express
  234. review_ticket["review-ticket"] --> fetch_ticket
  235. review_ticket --> express
  236. review_ticket --> simple
  237. maintainer_review --> simple
  238. check_impl --> simple
  239. manual --> simple
  240. review_ticket -. when available .-> verify_understanding["verify-understanding"]
  241. review_ticket -. when available .-> refine
  242. verify_understanding -. when available .-> refine
  243. verify_understanding -. when available .-> review_ticket
  244. review_ticket -. when available .-> check_impl
  245. review_ticket -. when available .-> attach
  246. check_impl["check-ticket-implementation"] --> fetch_ticket
  247. refine -. when available .-> manual["create-manual-test-instructions"]
  248. refine -. when available .-> plan["create-implementation-plan"]
  249. refine -. when available .-> express
  250. refine -. when available .-> attach
  251. plan -. when available .-> split["split-plan-tasks"]
  252. split --> execute["execute-plan-tasks"]
  253. execute -. when available .-> self_review
  254. execute -. when available .-> plan
  255. execute -. when available .-> harden
  256. self_review -. when available .-> handover["handover"]
  257. handover --> express
  258. express --> simple
  259. simple_rule["write-simple-explanations rule"] --> simple["explain-in-simple-language"]
  260. refine -. when available .-> simple
  261. plan -. when available .-> simple
  262. refine_pr -. when available .-> simple
  263. verify_understanding -. when available .-> simple
  264. split -. when available .-> simple
  265. execute -. when available .-> simple
  266. compact -. when available .-> simple
  267. memory_doctor -. when available .-> simple
  268. agentify["agentify-project"] -. when available .-> simple
  269. agentify -. when available .-> self
  270. grill_me["grill-me"] -. when available .-> simple
  271. self_review -. when available .-> simple
  272. agentify -. when available .-> compact_docs
  273. agentify -. when available .-> memory_doctor
  274. agentify -. when available .-> context_checkup["context-checkup"]
  275. context_checkup -. when available .-> agentify
  276. context_checkup -. when available .-> memory_doctor
  277. self_rule["self-improve-on-correction rule"] --> self["self-improve"]
  278. self --> compact["compact-skill-creator"]
  279. self --> compact_docs["compact-docs-writer"]
  280. compact --> compact_docs
  281. memory_doctor["memory-doctor"] --> self
  282. extract_pr["extract-pr-review-rules"] --> self
  283. refine_pr -. when available .-> self
  284. compact_gov["compact-governing-docs rule"] --> compact
  285. compact_gov --> compact_docs
  286. plans_rule["plans-directory rule"] -. informs .-> fetch_ticket
  287. plans_rule -. informs .-> attach
  288. plans_rule -. informs .-> fetch_pr
  289. plans_rule -. informs .-> refine_pr
  290. plans_rule -. informs .-> refine
  291. plans_rule -. informs .-> manual
  292. plans_rule -. informs .-> plan
  293. plans_rule -. informs .-> review_ticket
  294. plans_rule -. informs .-> check_impl
  295. plans_rule -. informs .-> handover
  296. plans_rule -. informs .-> self_review
  297. plans_rule -. informs .-> extract_pr
  298. docs_rule["self-contained-docs rule"] --> prepare_prompt["prepare-prompt"]
  299. docs_rule -. informs .-> fetch_ticket
  300. docs_rule -. informs .-> fetch_pr
  301. docs_rule -. informs .-> refine_pr
  302. docs_rule -. informs .-> refine
  303. docs_rule -. informs .-> manual
  304. docs_rule -. informs .-> plan
  305. ```
  306. ## Contributing
  307. Any agent works: the repo keeps one canonical [AGENTS.md](./AGENTS.md) that every agent reads, and
  308. [docs/contributor-setup.md](./docs/contributor-setup.md) explains the layout and how to hook yours
  309. up.

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.